Accounting for the uncertainty of speech estimates in the complex domain for minimum mean square error speech enhancement.
Ramón Fernandez AstudilloDorothea KolossaReinhold OrglmeisterPublished in: INTERSPEECH (2009)
Keyphrases
- speech enhancement
- signal to noise ratio
- minimum mean square error
- complex domains
- noise reduction
- speech signal
- reinforcement learning
- domain theory
- noisy environments
- edge detection
- bayesian networks
- confidence intervals
- denoising algorithm
- linear prediction
- wiener filter
- vocal tract
- edge preserving
- speech recognition
- higher order
- single channel
- motion estimation
- training data